Fix a deadlock with redo-oob.

If a checksummed target A used to exist but is now missing, and we tried to
redo-ifchange that exact file, we would unnecessarily run 'redo-oob A A';
that is, we have to build A in order to determine if A needs to be built.

The sub-targets of redo-oob aren't run with REDO_UNLOCKED, so this would
deadlock instantly.

Add an assertion to redo-oob to ensure we never try to redo-ifchange the
primary target (thus converting the deadlock into an exception).  And skip
doing redo-oob when the target is already the same as the thing we have to
check.

redo-ifchange.py

@@ -95,7 +95,8 @@ def should_build(t):
     f = state.File(name=t)
     if f.is_failed():
         raise builder.ImmediateReturn(32)
-    return dirty_deps(f, depth = '', max_changed = vars.RUNID)
+    dirty = dirty_deps(f, depth = '', max_changed = vars.RUNID)
+    return dirty==[f] and DIRTY or dirty
 
 
 rv = 202

redo-oob.py

@@ -10,6 +10,9 @@ if len(sys.argv[1:]) < 2:
 target = sys.argv[1]
 deps = sys.argv[2:]
 
+for d in deps:
+    assert(d != target)
+
 me = state.File(name=target)
 
 os.environ['REDO_NO_OOB'] = '1'

state.py

@@ -262,6 +262,7 @@ class File(object):
         reldep = relpath(dep, vars.BASE)
         debug2('add-dep: %r < %s %r\n' % (self.name, mode, reldep))
         assert(src.name == reldep)
+        assert(self.id != src.id)
         _write("insert or replace into Deps "
                "    (target, mode, source) values (?,?,?)",
                [self.id, mode, src.id])

t/stamp/stamptest.do

@@ -56,3 +56,9 @@ redo-ifchange usestamp usestamp2
 [ "$(wc -l <stampy.log)" -eq 5 ] || exit 71
 [ "$(wc -l <usestamp.log)" -eq 2 ] || exit 72
 [ "$(wc -l <usestamp2.log)" -eq 1 ] || exit 73
+
+# this simple test used to cause a deadlock.
+../flush-cache.sh
+rm -f stampy
+redo-ifchange stampy
+[ "$(wc -l <stampy.log)" -eq 6 ] || exit 74